linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Andy Lutomirski <luto@kernel.org>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	Fenghua Yu <fenghua.yu@intel.com>,
	Tony Luck <tony.luck@intel.com>,
	Yu-cheng Yu <yu-cheng.yu@intel.com>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	Borislav Petkov <bp@suse.de>,
	Peter Zijlstra <peterz@infradead.org>,
	Kan Liang <kan.liang@linux.intel.com>
Subject: [patch V3 63/66] x86/fpu/signal: Split out the direct restore code
Date: Fri, 18 Jun 2021 16:19:26 +0200	[thread overview]
Message-ID: <20210618143451.562513044@linutronix.de> (raw)
In-Reply-To: 20210618141823.161158090@linutronix.de

Prepare for smarter failure handling of the direct restore.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/fpu/signal.c |  110 +++++++++++++++++++++----------------------
 1 file changed, 56 insertions(+), 54 deletions(-)

--- a/arch/x86/kernel/fpu/signal.c
+++ b/arch/x86/kernel/fpu/signal.c
@@ -249,10 +249,7 @@ sanitize_restored_user_xstate(union fpre
 	}
 }
 
-/*
- * Restore the FPU state directly from the userspace signal frame.
- */
-static int restore_fpregs_from_user(void __user *buf, u64 xrestore, bool fx_only)
+static int restore_hwregs_from_user(void __user *buf, u64 xrestore, bool fx_only)
 {
 	if (use_xsave()) {
 		u64 init_bv = xfeatures_mask_uabi() & ~xrestore;
@@ -273,6 +270,56 @@ static int restore_fpregs_from_user(void
 	}
 }
 
+static int restore_fpregs_from_user(void __user *buf, u64 xrestore, bool fx_only)
+{
+	struct fpu *fpu = &current->thread.fpu;
+	int ret ;
+
+	fpregs_lock();
+	pagefault_disable();
+	ret = restore_hwregs_from_user(buf, xrestore, fx_only);
+	pagefault_enable();
+
+	if (unlikely(ret)) {
+		/*
+		 * The above did an FPU restore operation, restricted to
+		 * the user portion of the registers, and failed, but the
+		 * microcode might have modified the FPU registers
+		 * nevertheless.
+		 *
+		 * If the FPU registers do not belong to current, then
+		 * invalidate the FPU register state otherwise the task
+		 * might preempt current and return to user space with
+		 * corrupted FPU registers.
+		 *
+		 * In case current owns the FPU registers then no further
+		 * action is required. The fixup in the slow path will
+		 * handle it correctly.
+		 */
+		if (test_thread_flag(TIF_NEED_FPU_LOAD))
+			__cpu_invalidate_fpregs_state();
+		fpregs_unlock();
+		return ret;
+	}
+
+	/*
+	 * Restore supervisor states: previous context switch etc has done
+	 * XSAVES and saved the supervisor states in the kernel buffer from
+	 * which they can be restored now.
+	 *
+	 * We cannot do a single XRSTORS here - which would be nice -
+	 * because the rest of the FPU registers are being restored from a
+	 * user buffer directly. The single XRSTORS happens below, when the
+	 * user buffer has been copied to the kernel one.
+	 */
+	if (test_thread_flag(TIF_NEED_FPU_LOAD) && xfeatures_mask_supervisor())
+		os_xrstor(&fpu->state.xsave, xfeatures_mask_supervisor());
+
+	fpregs_mark_activate();
+	fpregs_unlock();
+	return 0;
+}
+
 static int __fpu_restore_sig(void __user *buf, void __user *buf_fx,
 			     bool ia32_fxstate)
 {
@@ -297,61 +344,16 @@ static int __fpu_restore_sig(void __user
 		user_xfeatures = fx_sw_user.xfeatures;
 	}
 
-	if (!ia32_fxstate) {
+	if (likely(!ia32_fxstate)) {
 		/*
 		 * Attempt to restore the FPU registers directly from user
-		 * memory. For that to succeed, the user access cannot cause
-		 * page faults. If it does, fall back to the slow path below,
-		 * going through the kernel buffer with the enabled pagefault
-		 * handler.
+		 * memory. For that to succeed, the user access cannot cause page
+		 * faults. If it does, fall back to the slow path below, going
+		 * through the kernel buffer with the enabled pagefault handler.
 		 */
-		fpregs_lock();
-		pagefault_disable();
 		ret = restore_fpregs_from_user(buf_fx, user_xfeatures, fx_only);
-		pagefault_enable();
-		if (!ret) {
-
-			/*
-			 * Restore supervisor states: previous context switch
-			 * etc has done XSAVES and saved the supervisor states
-			 * in the kernel buffer from which they can be restored
-			 * now.
-			 *
-			 * We cannot do a single XRSTORS here - which would
-			 * be nice - because the rest of the FPU registers are
-			 * being restored from a user buffer directly. The
-			 * single XRSTORS happens below, when the user buffer
-			 * has been copied to the kernel one.
-			 */
-			if (test_thread_flag(TIF_NEED_FPU_LOAD) &&
-			    xfeatures_mask_supervisor()) {
-				os_xrstor(&fpu->state.xsave,
-					  xfeatures_mask_supervisor());
-			}
-			fpregs_mark_activate();
-			fpregs_unlock();
+		if (likely(!ret))
 			return 0;
-		}
-
-		/*
-		 * The above did an FPU restore operation, restricted to
-		 * the user portion of the registers, and failed, but the
-		 * microcode might have modified the FPU registers
-		 * nevertheless.
-		 *
-		 * If the FPU registers do not belong to current, then
-		 * invalidate the FPU register state otherwise the task might
-		 * preempt current and return to user space with corrupted
-		 * FPU registers.
-		 *
-		 * In case current owns the FPU registers then no further
-		 * action is required. The fixup below will handle it
-		 * correctly.
-		 */
-		if (test_thread_flag(TIF_NEED_FPU_LOAD))
-			__cpu_invalidate_fpregs_state();
-
-		fpregs_unlock();
 	} else {
 		/*
 		 * For 32-bit frames with fxstate, copy the fxstate so it can


  parent reply	other threads:[~2021-06-18 15:25 UTC|newest]

Thread overview: 133+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-18 14:18 [patch V3 00/66] x86/fpu: Spring cleaning and PKRU sanitizing Thomas Gleixner
2021-06-18 14:18 ` [patch V3 01/66] x86/fpu: x86/fpu: Preserve supervisor states in sanitize_restored_user_xstate() Thomas Gleixner
2021-06-19  8:34   ` Borislav Petkov
2021-06-22 11:45   ` [tip: x86/urgent] " tip-bot2 for Thomas Gleixner
2021-06-18 14:18 ` [patch V3 02/66] x86/fpu: Make init_fpstate correct with optimized XSAVE Thomas Gleixner
2021-06-18 20:41   ` Thomas Gleixner
2021-06-18 20:44     ` Borislav Petkov
2021-06-22 11:45   ` [tip: x86/urgent] " tip-bot2 for Thomas Gleixner
2021-06-18 14:18 ` [patch V3 03/66] x86/fpu: Fix copy_xstate_to_kernel() gap handling Thomas Gleixner
2021-06-19  9:41   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 04/66] x86/pkeys: Revert a5eff7259790 ("x86/pkeys: Add PKRU value to init_fpstate") Thomas Gleixner
2021-06-18 14:18 ` [patch V3 05/66] x86/fpu: Mark various FPU states __ro_after_init Thomas Gleixner
2021-06-18 14:18 ` [patch V3 06/66] x86/fpu: Make xfeatures_mask_all __ro_after_init Thomas Gleixner
2021-06-20  8:44   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 07/66] x86/fpu: Get rid of fpu__get_supported_xfeatures_mask() Thomas Gleixner
2021-06-20  9:02   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 08/66] x86/fpu: Remove unused get_xsave_field_ptr() Thomas Gleixner
2021-06-18 14:18 ` [patch V3 09/66] x86/fpu: Move inlines where they belong Thomas Gleixner
2021-06-18 14:18 ` [patch V3 10/66] x86/fpu: Limit xstate copy size in xstateregs_set() Thomas Gleixner
2021-06-18 14:18 ` [patch V3 11/66] x86/fpu: Sanitize xstateregs_set() Thomas Gleixner
2021-06-20 21:30   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 12/66] x86/fpu: Reject invalid MXCSR values in copy_kernel_to_xstate() Thomas Gleixner
2021-06-21 10:00   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 13/66] x86/fpu: Simplify PTRACE_GETREGS code Thomas Gleixner
2021-06-18 14:18 ` [patch V3 14/66] x86/fpu: Rewrite xfpregs_set() Thomas Gleixner
2021-06-21 10:20   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 15/66] x86/fpu: Fail ptrace() requests that try to set invalid MXCSR values Thomas Gleixner
2021-06-21 10:17   ` Thomas Gleixner
2021-06-18 14:18 ` [patch V3 16/66] x86/fpu: Clean up fpregs_set() Thomas Gleixner
2021-06-21 12:05   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 17/66] x86/fpu: Make copy_xstate_to_kernel() usable for [x]fpregs_get() Thomas Gleixner
2021-06-21 12:32   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 18/66] x86/fpu: Use copy_xstate_to_uabi_buf() in xfpregs_get() Thomas Gleixner
2021-06-18 14:18 ` [patch V3 19/66] x86/fpu: Use copy_xstate_to_uabi_buf() in fpregs_get() Thomas Gleixner
2021-06-18 14:18 ` [patch V3 20/66] x86/fpu: Remove fpstate_sanitize_xstate() Thomas Gleixner
2021-06-18 14:18 ` [patch V3 21/66] x86/fpu/regset: Move fpu__read_begin() into regset Thomas Gleixner
2021-06-21 13:26   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 22/66] x86/fpu: Move fpu__write_begin() to regset Thomas Gleixner
2021-06-21 15:30   ` Borislav Petkov
2021-06-21 20:15     ` Thomas Gleixner
2021-06-18 14:18 ` [patch V3 23/66] x86/fpu: Get rid of using_compacted_format() Thomas Gleixner
2021-06-18 14:18 ` [patch V3 24/66] x86/kvm: Avoid looking up PKRU in XSAVE buffer Thomas Gleixner
2021-06-18 14:18 ` [patch V3 25/66] x86/fpu: Cleanup arch_set_user_pkey_access() Thomas Gleixner
2021-06-18 14:18 ` [patch V3 26/66] x86/fpu: Get rid of copy_supervisor_to_kernel() Thomas Gleixner
2021-06-18 14:18 ` [patch V3 27/66] x86/fpu: Rename copy_xregs_to_kernel() and copy_kernel_to_xregs() Thomas Gleixner
2021-06-21 18:00   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 28/66] x86/fpu: Rename copy_user_to_xregs() and copy_xregs_to_user() Thomas Gleixner
2021-06-21 19:44   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 29/66] x86/fpu: Rename fxregs related copy functions Thomas Gleixner
2021-06-21 23:00   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 30/66] x86/fpu: Rename fregs " Thomas Gleixner
2021-06-22  9:31   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 31/66] x86/fpu: Rename xstate copy functions which are related to UABI Thomas Gleixner
2021-06-22  9:38   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 32/66] x86/fpu: Deduplicate copy_uabi_from_user/kernel_to_xstate() Thomas Gleixner
2021-06-22 10:09   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 33/66] x86/fpu: Rename copy_fpregs_to_fpstate() to save_fpregs_to_fpstate() Thomas Gleixner
2021-06-22 10:16   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 34/66] x86/fpu: Get rid of the FNSAVE optimization Thomas Gleixner
2021-06-22 10:36   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 35/66] x86/fpu: Rename copy_kernel_to_fpregs() to restore_fpregs_from_kernel() Thomas Gleixner
2021-06-22 10:46   ` Borislav Petkov
2021-06-18 14:18 ` [patch V3 36/66] x86/fpu: Rename initstate copy functions Thomas Gleixner
2021-06-22 10:48   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 37/66] x86/fpu: Rename "dynamic" XSTATEs to "independent" Thomas Gleixner
2021-06-22 11:05   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 38/66] x86/fpu/xstate: Sanitize handling of independent features Thomas Gleixner
2021-06-21 20:42   ` Liang, Kan
2021-06-22 11:32   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 39/66] x86/pkeys: Move read_pkru() and write_pkru() Thomas Gleixner
2021-06-18 14:19 ` [patch V3 40/66] x86/fpu: Rename and sanitize fpu__save/copy() Thomas Gleixner
2021-06-22 13:51   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 41/66] x86/cpu: Sanitize X86_FEATURE_OSPKE Thomas Gleixner
2021-06-22 14:15   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 42/66] x86/pkru: Provide pkru_get_init_value() Thomas Gleixner
2021-06-18 14:19 ` [patch V3 43/66] x86/pkru: Provide pkru_write_default() Thomas Gleixner
2021-06-22 14:30   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 44/66] x86/cpu: Write the default PKRU value when enabling PKE Thomas Gleixner
2021-06-18 14:19 ` [patch V3 45/66] x86/fpu: Use pkru_write_default() in copy_init_fpstate_to_fpregs() Thomas Gleixner
2021-06-18 14:19 ` [patch V3 46/66] x86/fpu: Rename fpu__clear_all() to fpu_flush_thread() Thomas Gleixner
2021-06-22 14:40   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 47/66] x86/fpu: Clean up the fpu__clear() variants Thomas Gleixner
2021-06-22 15:38   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 48/66] x86/fpu: Rename __fpregs_load_activate() to fpregs_restore_userregs() Thomas Gleixner
2021-06-22 15:40   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 49/66] x86/fpu: Move FXSAVE_LEAK quirk info __copy_kernel_to_fpregs() Thomas Gleixner
2021-06-22 15:58   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 50/66] x86/fpu: Rename xfeatures_mask_user() to xfeatures_mask_uabi() Thomas Gleixner
2021-06-22 16:02   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 51/66] x86/fpu: Dont restore PKRU in fpregs_restore_userspace() Thomas Gleixner
2021-06-22 16:08   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 52/66] x86/fpu: Add PKRU storage outside of task XSAVE buffer Thomas Gleixner
2021-06-22 16:55   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 53/66] x86/fpu: Hook up PKRU into ptrace() Thomas Gleixner
2021-06-22 17:01   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 54/66] x86/fpu: Mask PKRU from kernel XRSTOR[S] operations Thomas Gleixner
2021-06-22 17:07   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 55/66] x86/fpu: Remove PKRU handling from switch_fpu_finish() Thomas Gleixner
2021-06-22 17:08   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 56/66] x86/fpu: Dont store PKRU in xstate in fpu_reset_fpstate() Thomas Gleixner
2021-06-22 17:10   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 57/66] x86/pkru: Remove xstate fiddling from write_pkru() Thomas Gleixner
2021-06-22 17:15   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 58/66] x86/fpu: Mark init_fpstate __ro_after_init Thomas Gleixner
2021-06-18 14:19 ` [patch V3 59/66] x86/fpu/signal: Move initial checks into fpu__sig_restore() Thomas Gleixner
2021-06-22 17:35   ` Borislav Petkov
2021-06-22 18:38     ` Thomas Gleixner
2021-06-22 18:43       ` Thomas Gleixner
2021-06-18 14:19 ` [patch V3 60/66] x86/fpu/signal: Remove the legacy alignment check Thomas Gleixner
2021-06-22 17:40   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 61/66] x86/fpu/signal: Sanitize the xstate check on sigframe Thomas Gleixner
2021-06-18 21:02   ` Andrew Cooper
2021-06-18 23:49     ` Thomas Gleixner
2021-06-22 19:10   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 62/66] x86/fpu/signal: Sanitize copy_user_to_fpregs_zeroing() Thomas Gleixner
2021-06-23  7:51   ` Borislav Petkov
2021-06-18 14:19 ` Thomas Gleixner [this message]
2021-06-23  8:10   ` [patch V3 63/66] x86/fpu/signal: Split out the direct restore code Borislav Petkov
2021-06-18 14:19 ` [patch V3 64/66] x86/fpu: Return proper error codes from user access functions Thomas Gleixner
2021-06-23  8:30   ` Borislav Petkov
2021-06-23 10:56     ` Thomas Gleixner
2021-06-18 14:19 ` [patch V3 65/66] x86/fpu/signal: Handle #PF in the direct restore path Thomas Gleixner
2021-06-23  8:45   ` Borislav Petkov
2021-06-18 14:19 ` [patch V3 66/66] x86/fpu/signal: Let xrstor handle the features to init Thomas Gleixner
2021-06-23  8:56   ` Borislav Petkov
2021-06-21 16:15 ` [patch V3 00/66] x86/fpu: Spring cleaning and PKRU sanitizing Yu, Yu-cheng
2021-06-21 21:53 ` Fenghua Yu
2021-06-21 22:22 ` Bae, Chang Seok
2021-06-22 20:02   ` Bae, Chang Seok
2021-06-24  0:06     ` Bae, Chang Seok
2021-06-22  1:59 ` Oliver Sang
2021-06-22  9:08   ` Thomas Gleixner
2021-06-22 18:56 ` Dey, Megha

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210618143451.562513044@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=bigeasy@linutronix.de \
    --cc=bp@suse.de \
    --cc=dave.hansen@linux.intel.com \
    --cc=fenghua.yu@intel.com \
    --cc=kan.liang@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=peterz@infradead.org \
    --cc=tony.luck@intel.com \
    --cc=yu-cheng.yu@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).